Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=tcld20

Climate and Development

ISSN: 1756-5529 (Print) 1756-5537 (Online) Journal homepage: https://www.tandfonline.com/loi/tcld20

Gender, environment and migration in Bangladesh

Kathinka Fossum Evertsen & Kees van der Geest

To cite this article: Kathinka Fossum Evertsen & Kees van der Geest (2019):Gender, environment and migration in Bangladesh, Climate and Development, DOI:10.1080/17565529.2019.1596059

To link to this article: https://doi.org/10.1080/17565529.2019.1596059

Published online: 08 Apr 2019.

Submit your article to this journal

Article views: 49

View Crossmark data

Gender, environment and migration in BangladeshKathinka Fossum Evertsena* and Kees van der Geestb

aParis School of International Affairs (PSIA), SciencesPo, Paris, France; bInstitute for Environment and Human Security, United Nations University, Bonn,Germany

ABSTRACTThis article addresses how gender norms impact the process of migration, and what this means for the useof migration as an adaptation strategy to cope with environmental stressors. Data was collected throughqualitative fieldwork, taking the form of semi-structured and open-ended interviews and focus groupdiscussions from a Dhaka slum and three villages in Southern Bangladesh’s Bhola district. Our datarevealed that women migrate when environmental stress threatens livelihoods and leave malehousehold members unable to earn enough income for their families. Employing an analyticalframework that focuses on the perceptions of individuals, this article shows how gender norms createsocial costs for women who migrate. Women thus have ambivalent feelings about migration. On theone hand, they do not wish to migrate, taking on a double work load, forsaking their purdah, andfacing the stigma that follows. On the other hand, women see migration as a means to help theirfamilies, and live a better life. While social costs negatively affect the utilization and efficiency offemale migration as an adaptation strategy to environmental stressors, it becomes clear that femalemigration is imperative to sustain livelihoods within the Bhola community.

ARTICLE HISTORYReceived 30 September 2017Accepted 22 February 2019

KEYWORDSGender; environment;climate change; migration;adaptation

1. Introduction

The rising acknowledgment of negative consequences of cli-mate change (IPCC, 2014; Warner, Van der Geest, & Kreft,2013), has led to an increased interest in the relationshipbetween human migration and the environment. Within thisexpanding academic field, migration is increasingly perceivedas a potential adaptation strategy to more intense and frequentenvironmental stressors (Black, Bennett, Thomas, & Bedding-ton, 2011; Tacoli, 2009). Among factors that shape migrationpatterns and experiences, gender roles are ‘perhaps one of, ifnot the single most important factor shaping the migratoryexperience’ (IOM, 2009).

The study tries to answer the following question: How doesgender influence the process of migration, and the utilizationand efficiency of migration as an adaptation strategy to environ-mental stressors?

The article employs an analytical framework which focuseson how individuals perceive different adaptation alternativeswhen environmental stressors threaten their livelihood. Itexplores how gender norms influence the perceptions of differ-ent options, and what alternative is ultimately chosen. Based onevidence from a local case study in Bangladesh the article showsthat, like men, women also migrate when environmental stres-sors impoverish the livelihoods of their households. Whilefemale migration is an important source of income for house-holds vulnerable to the negative effects of environmental stres-sors, perceptions of appropriate gender roles negatively affectopportunities and outcomes of female migration.

The article is structured in the following way. The first partinvestigates how women have been portrayed in the migration

literature over time; how migration is increasingly seen as anadaptation to environmental stressors; and the role of socialand gender norms in people’s adaptation choices. The secondpart reports on empirical fieldwork conducted in Bangladesh.In the last part of the paper findings are discussed and the con-clusion is presented.

2. Literature review

2.1. Female migration

In migration literature, female migrants were long portrayed aslargely passive, migrating for the purpose of marriage or familyreunification. Research conducted from the 1970s onwardschallenged this narrative, and pushed for the inclusion ofwomen as active agents in migration research (see for example,Chant, 1992; Curran, Shafer, Donato, & Garip, 2006; Pedraza,1991). Literature increasingly recognizes that women migratefor many reasons, of which marriage is only one. It has further-more been shown that more women migrate independently,and that migration for work is on the increase, both acrossand within borders (Deshingkar, 2005; Gosh, 2009, p. 8; Mar-tin, 2003, p. 4; Tacoli & Mabala, 2010).

Literature on women moving independently, moving with-out the rest of their household members, in South-Asia issparse. Also, migration statistics in this region tend to be ofpoor quality, and may misrepresent actual female migrationflows (Zlotnik, 2003). Despite this dearth of quantitative data,there has been an emergence of studies focusing on internalfemale labour migration in South-Asia (Deshingkar, 2005;IOM, 2009; Mazumdar & Agnihotri, 2014; Sundari, 2005),

© 2019 Informa UK Limited, trading as Taylor & Francis Group

CONTACT Kathinka Fossum Evertsen kathinka.f.evertsen@nord.no Faculty of Social Sciences, Nord University, Universitetsaleen 11, Bodø 8026, Norway*Present address: Faculty of Social Sciences, Nord University, Bodø, Norway.This article has been republished with minor changes. These changes do not impact the academic content of the article.

CLIMATE AND DEVELOPMENThttps://doi.org/10.1080/17565529.2019.1596059

showing that: ‘autonomous migration by women for employ-ment (…) is a phenomenon that can no longer be ignored’(Mazumdar & Agnihotri, 2014, p. 146). In the case of Bangla-desh, several studies show a substantial increase in femalemigration from rural to urban areas for wage work (Afsar,2002; Huq-Hussain, 1995; Kabeer, 1991).

Although first mentioned in the 1970s, the field of environ-mental migration only developed as a distinct research topic inthe early 2000s (Renaud, Bogardi, Dun, &Warner, 2007, p. 11).Given its youth and cross-disciplinary dynamics, it cannot beexpected that all topics related to the broad academic umbrellaof environmental migration have yet been covered. It is, how-ever, clear that the concept of actively migrating women,women migrating independently or otherwise taking an activepart in the migration process, has thus far been largely ignoredwithin environmental migration literature.

2.2. Gendered vulnerability and environmentalmigration

In view of the renewed interest in the relationship betweenenvironment and migration (Hunter, 2005; Piguet, 2013), it issurprising that the potential linkage between female migrationand environmental stressors has not been explored to a largerdegree (Chindarkar, 2012, p. 2). This missing link is especiallyprominent in the context of South-Asia, an area highly prone toclimate change and environmental stress – factors expected tosignificantly increase human mobility over the next decades.

A partial explanation for the lack of focus on women inenvironmental migration studies can be found in disaster andhazard literature where the vulnerability of women is comparedto that of men with disheartening results (see for example Can-non, 2002). Such studies have established that women are morevulnerable to environmental stress because of their subordinatestatus at home and in society (Ahmed & Fabjer, 2009; Björnberg& Hansson, 2013; Fothergill, 1996; Ikeda, 1995; Nelson, Mea-dows, Cannon, Morton, & Martin, 2002). Disproportionate vul-nerability of women is prominent in South-Asia (Cannon, 2002),ultimately leading to higher mortality rates for women in disastersituations. Although of crucial importance, this focus on vulner-ability can lead to victimization of women. It risks causing ablindness to the potential of female agency, failing to capturethe whole story (Jolly & Reeves, 2005, p. 2; Pedraza, 1991,p. 304). In line with the narrative of passive women, genderedmigration research concerning climate change impacts hasthus far focused on situations where the husband migrates andthe wife stays behind, leaving her to face increased hardshipsbecause of the absence of a male guardian. There are thereforefew studies on women as migrants (Boyd & Grieco, 2003).

There seems to be a growing awareness of this gap, andother studies have now been conducted focusing on indepen-dently migrating women in contexts of environmental stress.Jungehülsing (2010) shows that, although more vulnerablebecause of societal disadvantages, Mexican women migrate insearch of work when environmental stress negatively affectthe economy of their home community. Similarly, Tacoli andMabala (2010) show how young women are using migrationas a tool to diversify their livelihoods in Mali, Nigeria, Tanza-nia, and Vietnam. The authors find that independent migration

is on the increase among young women in response to a short-age of land in their home communities combined withincreased income opportunities in urban areas due to genderedlabour markets. Sundari (2005) has noted the same trend ofincreasing numbers of young women migrating internally inIndia. She explains this by citing the recurring drought in thesending areas coupled with the creation of new income oppor-tunities in export-oriented industries in urban areas.

If women migrate for environmental reasons more infor-mation is needed on the circumstances under which they doso. Failure to acknowledge this group of environmentalmigrants may cause the research community to miss the fullarray of adaptive measures undertaken. In view of female vul-nerabilities highlighted in disaster literature, it can be expectedthat female migrants face different and perhaps greater chal-lenges than men throughout the migration process. Thus, a bet-ter understanding of the ability of women to use migration asan adaption strategy, in comparison to men, is needed (Junge-hülsing, 2010, p. 20).

2.3. Migration as adaptation

Environmental migration has increasingly been perceived assomething to be supported and facilitated for improved adap-tation, rather than as a failure to adapt to environmental stress(Tacoli, 2009). This view builds on the argument that peoplewho are able to move can diversify their livelihoods, as migrationprovides opportunities for alternative income sources (Blacket al., 2011; Foresight, 2011, p. 21; see also Martin et al., 2014).

To investigate when and how migration can function as anadaptation strategy to environmental stressors, it is importantto distinguish forced and voluntary migration. Walsham arguesthat: ‘If migration is both planned and voluntary, it can providea social safety net for loss of income’ (Walsham, 2010, p. 7, myitalics). By contrast, when migration is forced, it tends to under-mine livelihood security and worsen the situation for migrantsand their relatives at home (Van der Geest & Warner, 2015).These criteria – planning and choice – can be used to evaluatewhen migration is useful as an adaptation strategy.

To better understand the circumstances under whichmigration can function as an adaptation strategy, two aspectsof migration patterns in societies prone to environmental stressneed to be investigated: namely, drivers of migration as well asconstraints to undertake such action. The Foresight report onmigration and global environmental change identifiesfive differ-ent but interrelated drivers for migration: economic, social, pol-itical, demographic, and environmental (Foresight, 2011, p. 46).

A key point made in the Foresight report is that the environ-ment can function both as an independent driver while alsoreinforcing the effect of other drivers of migration. Thus, tounderstand environmental migration, one should not onlylook for the visible environmental drivers, but investigatehow these interact with and affect other drivers of migrationalready prominent in the community in question, as ‘substan-tial social, economic and human capital may be required toenable people to migrate’ (Foresight, 2011, p. 12).

In view of this argument, also constraints are important fac-tors determining migration. Adaptation constraints are definedby the International Panel for Change (IPCC), as ‘factors that

2 K. F. EVERTSEN AND K. VAN DER GEEST

make it harder to plan and implement adaptation actions’(Klein et al., 2014, p. 907). Ultimately, constraints can causepeople to be ‘trapped’, when they ‘need to move for theirown protection but… lack the ability’ (Black & Collyer,2014). Constraints to adaptation has often been categorizedas ecological, physical, economic or technologic (Adger et al.,2009, p. 337). Challenging this view, both Adger with others(2007) and Jones and Boyd (2011) have used the term ‘socialbarriers to adaptation’ to describe how also norms and valuesmay cause people to decide against what can be objectivelyviewed as the most optimal action for adaptation.

Social and cultural norms colour how potential migrantsperceive their different options by discerning what is appropri-ate behaviour. In accordance with what is perceived as appro-priate, different alternatives for actions will lead to differentlyvalued outcomes. Consequently, the perception of differentoptions influence what decisions are made and what actionsare carried out (or not) (Martin et al., 2014).

Gender, the social construct that guides what is appropriatebehaviour for men and women respectively (IOM, 2009, p. 10),is often highlighted as a factor expected to greatly influence anindividual’s migration behaviour. Understanding gender as asubjective process, where expectations of appropriate behav-iour for men and women are internalized and then acted outin interaction with others, gender can be expected to influencewhich actions are perceived as more or less desirable, and sub-sequently what is seen as normatively and practically possible(Adger et al., 2009; Jones & Boyd, 2011; Kabeer, 1991).

3. Analytical framework

The concept of social barriers to adaptation is useful for inves-tigating how gender roles can work as a constraint on femaleadaptation. However, we sought an analytical framework thatalso includes the possibility of individual agency and allows

for investigating how an individual negotiates both opportu-nities and constraints. The analytical framework of Grothmannand Patt (2005) is useful for this purpose. It considers whichopportunities are perceived to be accessible and how desirablethese opportunities are.

Grothmann and Patt divide the decision-making of adap-tation choices into two phases (2005, pp. 200–203). First, a‘risk appraisal’ is carried out, where the individual evaluatesthe probability of a threat and how harmful the potential con-sequences of this threat are. Next, if the threat is perceived to beboth likely and grave an ‘adaptation appraisal’ is carried out,where the individual evaluates his or her own capacity to actupon the perceived risk. The adaptation appraisal is dividedinto three subcomponents: (1) how efficient the adaptiveactions are thought to be, (2) whether the individual perceivesthe action as possible for him or her to carry out, and (3) theanticipated costs associated with this action. These three com-ponents form the perceived adaptive capacity of the individual.The authors stress that cognitive processes – how individualsperceive the world around them – may be as important asmaterial resources in influencing decision outcomes:

(…) if agents systematically underestimate their own ability toadapt, this qualifies as a more important ‘bottleneck’ for adaptationthan the objective physical, institutional or economic constraints.(Grothmann & Patt, 2005, p. 203).

Altering Grothman and Patt’s model to include gender as influ-encing people’s perception of alternative options, as shown inFigure 1, it can help shed light on how gendered norms andvalues in a community affect who migrates and why.

4. Methods

Given the study’s explorative nature and the need for an in-depth understanding of the actors’ motivations to comprehendthe effects of gender on migration patterns, a case study, using

Figure 1. Analytical framework.

CLIMATE AND DEVELOPMENT 3

qualitative tools, was deemed appropriate to answer theresearch question of this study (Gerring, 2004; Yin, 1994, p. 13).

Prior to the actual fieldwork, six interviews with academicspecialists on gender, climate change and development wereconducted. Insights from these interviews were used to designthe research tools. During the fieldwork, semi-structured andopen-ended interviews were conducted with 9 male and 17female respondents in the Bhola bustee and the Bhola district.In addition, we conducted four focus group discussions, twowith men and two with women. See Table 1 for overview. Inter-views with key informants, including local leaders, non-govern-mental organizations, and local government officials, wereconducted in all stages of the research process. See Annex Afor full list of respondents.

Bangladesh was chosen as a case in view of three of the coun-try’s characteristics. First, migration is already an important

means of income-diversification (Afsar, 2003; McNamara,Olson, & Rahman, 2016; Siddiqui, 2003). Second, Bangladeshhas been listed the fifth country most at risk from disasters,the risk index comprising exposure to natural hazards and thevulnerabilities of the society (Mucke, 2014, p. 9). Having nega-tive consequences of climate change threatening their liveli-hoods, a great number of Bangladeshis face a need to migrate.However, many lack the capacity to do so. Third, Bangladeshis a country where deep gender inequalities persist (Amin,1997, p. 213; Kabeer, Mahmud, & Tasneen, 2011, p. 7). Genderroles in Bangladesh are heavily influenced by the cultural prac-tice of purdah, which can be understood as ‘the broader set ofnorms and structures that set standards of female morality’(Amin, 1997, p. 213). Excluded from the public, the home natu-rally becomes women’s domain and affects the division of labouramong householdmembers. Men are required to provide for thefamily by generating income outside of the home, while womenare obliged to take care of domestic tasks (Kabeer et al., 2011,p. 7). It is assumed that the effect of gender on migration isstronger where gender inequalities are greater. In view of thestrength of purdah, Bangladesh is a country where one canexpect to see strong effects of gender on migration behaviour.

A high number of migrating women take up work in the gar-ments sector. Therefore, the first part of the fieldwork was con-ducted in the capital, Dhaka, where the majority of garmentfactories are located. Although many migrants settle elsewhere,a significant number of poor migrants settle in the slums in andaround cities (Walsham, 2010, p. 15). Furthermore, because oftheir low income, many garment workers are slum dwellers.The smaller geographical areas that slums occupy aided inidentifying female migrants. The Bhola bustee (bangla wordfor informal settlement or slum) particularly stood out. Theoverwhelming majority of its inhabitants are from the Bholadistrict in coastal Bangladesh, which is highly exposed toenvironmental stressors (McNamara et al., 2016). Peoplehave been moving to Dhaka from the Bhola district since the1971 cyclone, establishing a migration pattern between thetwo locations. Furthermore, the slum is located in the Mirpurarea of Dhaka, where numerous garment factories are located.

Next, fieldwork was carried out in sending communities inBhola island to get a better sense of this community’s attitudetoward migration in general and perceptions of male andfemale migrants in particular. Open-ended interviews andfocus group discussions were conducted in Ilisha Union Par-ishad, Dhania Union Parishad, and Syeddpur Union Parishadin the Bhola district, as indicated by the red (opened) circlesin Figure 2. The study sites were selected from the hazardmap displayed in Figure 3, showing riverbank erosion acrossthe Bhola district between 1973 and 2005 (Sarker & CEGISin Inmam, 2009).

Lack of quantitative data limits the scope of this study. TheBangladesh population census does not have detailed dataabout environmental migration disaggregated by sex. Suchdata would have been useful for a quantitative analysis of pat-terns and historical trends in female migration from environ-mental hotspots like the Bhola district. While additionalquantitative data collection could have benefited the results ofthis study, limited time and resources did not allow for suchdata gathering.

Table 1. Interviews and focus group discussions.

Study site Gender Interviewees Focus groups

Bhola Island Female 4 2Male 2 1

Bhola Bustee Female 13 0Male 7 1

Figure 2. Field sites in the Bhola district.

4 K. F. EVERTSEN AND K. VAN DER GEEST

5. Findings

5.1. Sending and receiving communities

Situated in the Bay of Bengal, the coastal Bhola district isexposed to the activity of two rivers, as well as to tidal changesand cyclones from the ocean. Of the many environmental stres-sors affecting the area, river bank erosion was identified by anoverwhelming majority of the respondents as the mainenvironmental stressor influencing their decision to move.

The main occupations in the Bhola district are fishing andfarming (Arzu, Mayors office Bhola zila, personal communi-cation, September 21, 2014). Fishermen represent the pooreststrata of society in the district, and live close to the riverwhere land is cheap. Poor fishing families are most affectedby river bank erosion, lacking resources to move further inland.The Meghna River strips soil from the island’s eastern bankand, in the process, destroys farmland and displaces families.The loss of farmland pushes farmers into fishing, an occupationwith which they have little experience.

Several respondents also complained that there is ‘less fish inthe river’, causing food insecurity large parts of the year. As fishis a fundamental natural resource to these households’ liveli-hoods, declining fish stocks further increases the need foralternative income strategies in the district. For women in theBhola district, no real income opportunities are available.

In response to the decrease in natural resources, many seemigration as the only way to find new income sources. Peoplemove to the larger cities like Chittagong and Dhaka to work asday labourers, in the garments, or as domestic workers,depending on their gender, age, and social networks.

Dhaka, the capital of Bangladesh, is swelling beyond itscapacity, which affects available income opportunities, as wellas living conditions (McNamara et al., 2016, p. 2). Furthermore,many low-lying cities are vulnerable to environmental stressors,and Dhaka is no exception (Adamo, 2010, p. 162; Simon, 2010;Adri & Simon, 2018). Consequently, many migrants find them-selves in a situation where they leave one set of problems behindfor a new set of vulnerabilities in the destination area (Ayeb-Karlsson, Van der Geest, Ahmed, Huq, & Warner, 2016; Fore-sight, 2011; McNamara et al., 2016). In the Bhola slum, threatsof eviction, fires, and flooding constitute constant stress for theslum-dwellers. Yet, for many rural poor, Dhaka is viewed as anopportunity for alternative income sources and a better life.

5.2. Women as migrants

In the Bhola district, when asked about female migration, everyvillager knew someone who had moved, often responding ‘yes,a lot, a lot’. Somewhat surprisingly, in several conversationswith villagers in Bhola, it was reported that almost equal

Figure 3. Hazard map showing riverbank erosion.

CLIMATE AND DEVELOPMENT 5

numbers of women and men migrate. Thus, we could establishan existing trend of female migration from the Bhola district.

In the receiving community in the Bhola slum in Dhaka, weidentified three categories of migrating women: (1) women whohad migrated with their husbands, (2) women who were thehead of their households, typically divorced or widowed, and(3) young unmarried women who had migrated alone to pro-vide for themselves and their families back in the village.

The agency exercised – by all three categories of migratingwomen – was a surprising finding. In view of the norm thatwomen need approval from their male guardians before leavingthe home, it was expected that female migrants would in someway or other have been ‘sent’ by their families. This assumptiondid not hold. With only a few exceptions, the female respon-dents took the initiative and convinced skeptical familymembers to allow them to leave, or to migrate with them. 17-year-old Anika explained: ‘I told them a lot of things to makethem understand’ (Bhola slum, 29.08.2014).

Women migrating independently are mostly younger,unmarried women. This is not surprising, and is found alsoin other studies, which show how married women are theleast likely to migrate (Afsar, 1994; Jungehülsing, 2010; Massey,Fischer, & Capoferro, 2006). (Afsar, 1994; Jungehülsing, 2010;Massey et al., 2006). Household responsibilities, and especiallythe obligation to take care of the children, make it difficult formarried women to leave their household behind. Leaving herchildren behind, a woman would be deemed a ‘bad mother’.Several female participants of whom had migrated withouttheir children, before returning to their village, expressedguilt in this regard (Bhola district, focus group discussion,26.09.2014). Alam, a male villager, explained: ‘basically nobodyleaves their children behind here. If they move to Dhaka they allgo along’ (Alam, Bhola district, 29.09.2014).

An important lesson taken away from the fieldwork is thatagency should not be ascribed only to independentlymigrating women, however. Although less visible, marriedwomen also exercise strong agency in the migration process.For example, the initiative to migrate is often taken by thewife. This was confirmed by a substantial share of the malerespondents, who explained how it had been a female house-hold member – mothers, sisters, and wives – who had firstsuggested that the household should migrate. After arrivingin Dhaka with their household, married women also take upwage-work alongside their husbands. Nahar, wife and mother,recalled:

I said, (…) ‘I am going to go to Dhaka so I can feed my kids. (…) Iwill go to work myself so we can eat.’ I said that and made himunderstand, and we came [to Dhaka] (Bhola slum, 04.09.2014).

5.3. Gendered opportunities

An interesting characteristic of the Mirpur area in Dhaka wherethe Bhola slum is located is that the labour market is segregatedby gender in a way that allows for a more stable income forwomen than for men.

Women in the Bhola slum are mostly occupied either ashousemaids or garment workers. While the income of house-maids is also of importance, it is the presence of numerous gar-ment factories that explains the favourable working

opportunities for women in this area, as the wages are signifi-cantly higher in the garments sector than in traditional house-hold work.

The garment factories do not exclusively employ women,but prefer to do so. In the past, more men also worked inthese factories, but this has changed due to harder competitionin the labour market (Evertsen, 2015, pp. 61–63). As a result,many men that used to work in the garments sector in Mirpurhave now lost their jobs. Slum-dwellers explained that employ-ers prefer to hire women because they are easier to control.Women will not complain over bad working conditions orlow salaries Rafiq explained: ‘[The employers] know womenwill not come forward to ask them for money. But the menwill. That’s why they work with the women’ (Bhola slum,01.09.2014).

Most of the male slum-dwellers work as day labourers,taking odd jobs wherever and whenever available. Typicalwork includes rickshaw pulling, cutting soil, constructionwork, garbage management, painting, and carpeting. Strongcompetition over few jobs only allows men to find workbetween 10 and 15 days a month, the slum-dweller Alamexplained (Bhola slum, 01.09.2014).

The fact that the garment factories prefer to employ womenover men has two contrary effects. Women’s subjugation is themain reason why they are the preferred workers, and theiremployment contributes to a reinforcement of this subjugationas women are often underpaid and mistreated in their work-place (Afsar, 2002, p. 106; HRW, 2015). At the same time,the gendered segregation of the labour market in Mirpur resultsin women often having a more stable income than men withinthe slum community. Men’s salaries are higher per hour, butthey tend to work less hours. Consequently, female income isof high importance to households in the Bhola slum.

This reliance on female income finds itself in striking dis-agreement with socially accepted gender norms in Bangladesh.Nevertheless, facing lack of income opportunities in the Bholadistrict, the garments industry creates an economic pull factorfor women to migrate, implying that women perceivemigration as an efficient way to cope with the risk that environ-mental stressors pose to livelihoods.

5.4. Social costs

There seems to be a consensus among researchers and develop-ment organizations that a job in the garments sector has ahigher status in society than jobs previously available towomen in Bangladesh, which has typically been domesticwork (Banks, 2013; Jansen, personal communication, February28, 2015; Nasreen, personal communication, August 7, 2014).Domestic work has been regarded as unsafe for the womenin question, often bearing the stigma of sexual harassment.To be a housemaid has furthermore never received the statusof a ‘real job’, as household tasks are already the responsibilityof women. Work in the garments sector is perceived as saferthan domestic work, making households more willing to givefemale members permission to work.

However, this study finds that women working in the gar-ments sector are also highly stigmatized, in some ways evenmore so than more traditional domestic workers. A recurrent

6 K. F. EVERTSEN AND K. VAN DER GEEST

explanation of why the garments sector is not suitable forwomen was that it will make them ‘go bad’. Investigatingwhat was meant by ‘bad’, it became evident that the negativestigma attached to ‘city girls’ is linked to the perception ofthem violating their purdah.

In view of the stronghold of purdah, the idea of workingwomen is contested in Bangladesh. While garment work is per-ceived as a ‘real job’ in a workplace wheremen also work, the factthat women and men work together is also the crux of the pro-blem with regard to female reputation, as it breaks the physicalsegregation of women and men. This will lead female garmentworkers to be viewed with suspicion, their behaviour closelyscrutinized. Respondents expressed concern that women work-ing long hours outside the home will not have time to properlydo their prayers, or be able to fully cover their bodies when doingphysical labour work, both seen as important elements ofupholding purdah. Not being able to cover properly will leadto increased attention by male coworkers, leading to accusationof the women being immoral and ‘loose’. The accusations ofwomen working in garments differ from the stigma associatedwith domestic workers, who are still confined within the wallsof the home and are largely portrayed as victims rather thanactive violators of the purdah regime.

To be perceived as a ‘bad’ woman has high social costs, bothfor the woman in question, and for her family members. Onesuch consequence is the unmarried women’s decreasing valueon the marriage market. Saving for one’s own dowry is a com-mon rationale for young women to migrate to Dhaka. How-ever, the price of dowry is closely linked to reputation. Thestigma associated with being a ‘city girl’ increases the amountof dowry to be paid to the groom’s family for acceptance ofthe marriage. This requires them to stay in Dhaka for a pro-longed period of time, potentially further increasing theirdowry as this also rises with age (Huda, 2006, p. 255).

Importantly, purdah is not a concept that only involveswomen. It is a relational social construct regulating specificresponsibilities and appropriate behaviour for both men andwomen. When a woman takes up wage work it indicates thather male guardian has failed both to provide for her and to pro-tect her, signalizing that he has failed to fulfil his responsibilitiesas a man. Tasfiq explained: ‘I obviously felt bad that [my sister]had to work’ (Bhola slum, 14.09.2014). Male honour thus cre-ates a substantial constraining factor for women who wish toenter the labour force.

Because taking care of home and children are at the core offemale responsibilities, male household members will oftenrequire an assurance that wage work will not compromisesuch tasks before permitting their women to earn a wage. ‘Itold her that if she can manage to work, then she should,’one male slum-dweller recalled (Reaj, Bhola slum,15.09.2014). A problem mothers faced, was that only a few fac-tories offer childcare services. Several female respondentsexplained that they worried for their children while at work,leaving them home alone or with neighbours: ‘I leave my sonhome alone […] I feel anxious. If he is ill, I can’t be nearhim, that’s what I feel bad about’ (Ayela, Bhola slum,05.09.2014). Having the responsibilities for home and children,going to work, and facing social stigma when doing so, womenin the Bhola slum face a triple burden.

It becomes clear that it is important for working migrantwomen – as well as for their household members – to upholda positive reputation to the extent possible. Such efforts concealactual conditions and hardships of the migrants, with thepotential consequence of further deepening the costs thatmigration entails. When asked about her visits to her home vil-lage Anika, a young garment worker in Dhaka shrugged: ‘I can’tsay it’s not nice, so I said it’s nice’ (Bhola slum, 29.08.2014).

We asked migrants what they had expected to earn beforecoming to Dhaka. Respondents often often answered a numberfour times higher than their actual salary. This high numberwas also reported by villagers when asked what people inDhaka earn. One consequence of such misinformation is thatmigrants see the need to stay in Dhaka longer than planned.Several respondents explained that they wanted to save enoughmoney to buy a new plot of land in the village, having lost landand home to river bank erosion. Because they earned less thananticipated, they were unable to return as planned. This wastrue for both male and female respondents. However, aswomen face larger social – and subsequent economic – costswhen migrating, the consequences of such misinformationmay be correspondingly more severe.

5.5. Justification

The male stigma associated with female migration leaves theinitiative to migrate for wage work with the women in question,simply because men will not ask their women to work. In viewof negative social stigma and subsequent economic lossattached to female migration, we wanted to explore howmigration nevertheless becomes a normatively justifiableoption for women in the Bhola district.

It became clear from conversations with female migrantsthat they put the blame of the violation of their purdah upontheir male household members, who, by failing to provide forthem, forced women to take up wage work. The female respon-dents perceived their purdah as valuable, being closely linked totheir identity and honour. At the same time, blaming men forthe violation of purdah also functioned as a justification forwomen to step out of the home, as their men left them withfew other options. Using hardship to justify their actions,these women seemed to have found a pragmatic balancebetween the norms of purdah and the reality of their situation:

Staying under purdah is very good, but you can’t sit back and onlythink about purdah and not do anything. (…) So you have to main-tain Allah’s rules and also have to lead your life (Jameela, Bhola dis-trict, focus group discussion, 27–28.09, 2014).

Similar to the reasoning behind men’s responsibilities towardwomen, female migrants often justified their actions by refer-ring to their responsibilities towards parents, younger siblings,or children. Anika, a young garment worker, explained how sheworked so she could stand with her ‘head held high’ (Bholaslum, 29.09.2014), justifying her working outdoors by empha-sizing the importance of supporting her younger siblings.

5.6. Resistance

While some women find justifications that allow them tomigrate, such rationalizations are challenged by women who

CLIMATE AND DEVELOPMENT 7

have not migrated. During the focus group discussion in Bholawith women not connected to the garments sector, a strikinglevel of anger was expressed toward migrating women. Onerespondent burst out: ‘We hate the garments!’ (Bhola district,focus group discussion, 28–29.09.2014). The other respondentsexpressed strong-felt agreement. The group associated clearnegative characteristics with women migrating for wagework. They explained how such women are ‘greedy’, thinkingabout themselves rather than their families’ reputation.

This anger may have two different but interrelated expla-nations. First, the inappropriate behaviour of migratingwomen may feel like an attack on the self-respect of thosewho stayed behind. An alternative explanation is that womenremaining in Bhola envy the women who migrate for takingsuch an opportunity when they are themselves equally dis-tressed. Still, they did not reach the same conclusion as thewomen who chose to migrate. Rather, their rationale seemedto be that it is their responsibility as women to keep the house-hold together, no matter the situation. If a woman migrates, itsymbolizes a breakdown of the household: ‘Women are respon-sible for maintaining the family with husband’s little income’(Bhola district, focus group discussion, 27–28.09.2014), it wasexplained.

An apparent question is what differences exist betweenmigrating and non-migrating women that can explain theircontrasting decisions. One such difference may be educationallevel. While we asked respondents about their educationalbackground, most respondents were migrants, which gives uslittle basis for comparison with non-migrants. Earlier researchhas shown that people who migrate due to loss of livelihoodsare generally worse off and have less education than peoplemigrating for other reasons (Adri & Simons, 2018, p. 326).The Bhola district is poor, and most respondents only had afew years of schooling. It is therefore not certain that migratingwomen have more education than non-migrating women. Thiswould be an interesting point for further research.

6. Discussion

6.1. Risk and adaptation appraisals

According to the model for cultural perceptions laid out in sec-tion 3, an individual will carry out a ‘risk-appraisal’, evaluatingthe likelihood and magnitude of a threat, before evaluatingpotential routes for adaptation. For women in the Bhola dis-trict, the alternative adaptation routes identified in responseto river bank erosion are to move further inland or to migratefor wage work. In line with Grothman and Patt’s model, itseems that women only consider migration as a real alternativewhen they perceive it a high probability that their male guar-dian will not be able to provide for the household and thatthis will have serious consequences. When women identifythe need to provide for their household, some will migrate inspite of social and cultural norms discouraging them fromdoing so. As Nahar, mother of four, simply said: ‘My husbandworking alone can’t do it’ (Bhola slum, 04.09.2014).

Female migration as a response to lack of male incomeshould not be understood as an automatic process, however.As showed by Grothman and Patt, the existence of a high

risk with grave consequences is not enough to explain adaptiveaction. An adaptation appraisal also has to be concluded beforeaction is taken. In this case, the women will weigh how efficientthey think migration will be, whether it is physically and nor-matively possible for them to migrate, and what the costsassociated with migration are. If the risk is perceived as grow-ing, so may the likelihood for migration being perceived as areal alternative.

The gendered income opportunities, with the availability ofwork for women in the garments factories in Mirpur, makesmigration appear as an efficient migration strategy. However,this study has shown how the concept of working women ishighly contested in the Bhola community. Purdah is highlyvalued, and it is important for both men and women to upholdpurdah to the extent possible. Grothmann and Patt (2005) seemto be right in their argument that perceived adaptive capacitycan heavily influence adaptive behaviour. Values and normsheld by the community play an important role in an individ-ual’s decision making. Although the loss of livelihoods in thehome community and economic opportunities in Dhaka areimportant factors influencing the decision to migrate, thenotion of ‘self-respect’ is of outmost importance in this case,and is not easily given up for monetary goals. Indeed, some-times the adaptation appraisal carried out by individuals willconclude that the costs associated with migration are toohigh, and that moral reputation is of more value. For somewomen, it becomes normatively impossible to migrate.

6.2. Implications for adaptation

Interviews and focus group discussions conducted withmigrant and non-migrant men and women in the Bhola com-munity reveal that women have to weigh migration opportu-nities against higher social costs than men. Consequently, alarger risk may be needed before women perceive the risk assufficient for action, highlighting the interlinkage betweenrisk- and adaptation appraisals. If this holds, it means thatwhen affected by the same negative consequences of environ-mental stressors, women will wait longer before they migratethan men will.

This has negative implications for adaptation as it will renderhouseholds with more female household members more vulner-able to environmental stress. If migration is delayed until there isno other option left – for example when farmers are unable tomake a living because farmland is lost to river bank erosion-there is a risk that the migration will be carried out in a lessplanned manner. Several female respondents explained howthey had lived on the streets of Dhaka for some time beforefinding their way to the Bhola slum. In situations wherewomen do not have the support of their household uponmigrating, it is possible that they are also less able to utilizemigration networks that the household would otherwise drawsupport from. In short, the fact that women wait to migratewill expose women to greater risks both at home and in the des-tination area. Because it is regarded as unsafe for women to travelto Dhaka alone, some women also wait for a substantial amountof time after they have made the decision to migrate. Anikaexplained how she waited in the village for three years beforeshe could travel with a neighbour (Bhola slum, 29.08.2014).

8 K. F. EVERTSEN AND K. VAN DER GEEST

That women wait for years to migrate because of gender-related constraints shows that they may be temporarily trapped,until an opportunity to migrate emerges. Furthermore, not allwomen facing the same situation of livelihood stress migrate.Some wish to stay while others feel like they can not move.The informal settlements alongside the embankments indicatea significant trapped population in the area, as entire house-holds are being displaced. Women may not only be trappedin monetary terms, but also but also culturally within thewalls of their home. As one of the returned garment workersexpressed: ‘Girls stay in the house as if they stay in jail’ (Nas-reen, Bhola district, focus group discussion, 26.09.2014).

In section 2.3 we argued that planning and choice can beused to evaluate when migration is useful as an adaptationstrategy. Women in the Bhola community face challenges inboth these respects. Lacking choice of when to migrate andwhen to return, leave women unable to plan for their future,and affect how efficient female migration is as an adaptationstrategy to environmental stressors.

The fact that income opportunities are more stable forwomen than men in the destination area would, from a purelyeconomic perspective, be an argument in favour of more femalemigration. However, when considering the social costs, andtheir subsequent negative economic effects, the positive out-comes of migration diminishes. This study shows how femalemigration is an important adaptation tool, but that better facili-tation is crucial to avoid that women suffer from their choice.

7. Conclusions

The findings of this study support the idea that gender influ-ences all aspects of the migration process. Exercising stronginfluence on the behaviour of individuals within the house-hold, gender directs what income-generating opportunitiesare available, the level of wages, status of work, likelihoodof migration, and how migrants are perceived. Gendernorms also impact vulnerabilities by delaying adaptivebehaviour and compromise the economic gains from femalemigration.

In view of the large social costs associated with femalemigration, strong environmental and economic push factorsare required to explain the relatively large share of femalemigrants from the Bhola district. Yet, not all women facingthe same push-factors migrate. Having to negotiate economicincentives and cultural constraints women in the Bhola districtfeel ambivalent towards migration. On the one hand, they donot wish to migrate, taking on a double work load, forsakingtheir purdah, and facing the stigma that follows. On theother hand, women see migration as a way to help theirfamilies, and to be able to live a better life. Such dilemmasmake it difficult to determine when women are trapped,when they move voluntarily, and when they are forced to do so.

No one should be forced to migrate or forced to stay. Oneshould therefore be careful to conclude on behalf of womenin the Bhola district what would be the best way for them toadapt to environmental stressors. It is clear, however, that tobetter facilitate for women who need and want to migrate isimperative.

First, female environmental migration need to be brought tothe attention of development agencies, donors and policymakers. Increased awareness is a precondition for better facili-tation. Second, the main issue to address is the social stigmaassociated with working women. This study has shown thisstigma comes from the assumption that working women areunable to adhere to the practice of purdah. Thus, one solutioncould be to allow breaks for prayers and ensure temperaturesthat allow female workers to cover their bodies as they wantinside. Furthermore, providing day-care for children wouldhelp lessen the extra burden of work which women are facing.These could be important first steps to improve facilitation offemale migration from Bhola.

Women are more active agents in migration processes thanexpected. This study concludes that women should not be per-ceived as passive, but rather active agents in migration pro-cesses, who at the same time face greater struggles than menin achieving their goals. To better facilitate for planned andvoluntary migration as a way for vulnerable households toadapt to environmental stressors, and in view of the chal-lenges migrating women often face, there is a need for moreresearch on the interlinkages between gender and environ-mental migration. It is also clear that female environmentalmigrants deserve more attention than they have thus farreceived.

Acknowledgements

We want to extend our thanks to François Gemenne and Sonja Ayeb-Karlsson for advice and feedback during the research process; to SaleemulHuq, Sarder Shafiqul and the rest of the ICCCAD team for making the fieldwork possible; to Casey Williams for proof reading; and to everybody whohave taken the time to contribute information, comments and support,including Ana Namaki, Andreas Uhre and Kenneth Bo Nielsen. We ded-icate this article to the Bhola community, who gave of their time andshared their stories.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Kathinka Fossum Evertsen is a PhD Fellow at Nord University Faculty ofSocial Science and holds a master’s degree in Human Rights and Humani-tarian Action from Paris School of International Affairs (PSIA), Science-sPo, Paris. Kathinka’s research focuses on the gendered aspect ofenvironmentally induced human movement and adaptation.

Kees van der Geest (PhD) is a human geographer who studies the impactsof climate change, human mobility and livelihood adaptation from apeople-centred perspective. His work has contributed substantially to theempirical evidence base on migration-environment linkages and impactsof climate change beyond adaptation (’loss and damage’).

References

Adamo, S. B. (2010). Environmental migration and cities in the context ofglobal environmental change. Current Opinion in EnvironmentalSustainability, 2, 161–165. doi:10.1016/j.cosust.2010.06.005

Adger, N., Agrawala, S., Mirza, M., Conde, C., O’Brien, K., Pulhin, J.,…Takahashi, K. (2007). Assessment of adaptation practices, options, con-straints and capacity. In M. Perry, O. Canziani, J. Palutikof, P. van derLinden, & C. Hansen (Eds.), Climate change 2007: Impacts, adaptation

CLIMATE AND DEVELOPMENT 9

and vulnerability. Contribution of working group II to the fourth assess-ment report of the intergovernmental panel on climate change (pp. 717–743). Cambridge: Cambridge University Press.

Adger, N., Dessai, S., Goulden, M., Hulme, M., Lorenzoni, I., Nelson, D. R.,…Wreford, A. (2009). Are there social limits to adaptation to climatechange? Climatic Change, 93, 335–354. doi:10.1007/s10584-008-9520-z

Adri, N., & Simon, D. (2018). A tale of two groups: Focusing on the differ-ential vulnerability of “climate-induced” and “non-climate-induced”migrants in Dhaka city. Climate and Development, 10(4), 321–336.doi:10.1080/17565529.2017.1291402

Afsar, R. (1994). Internal migration and women: An insight into causes,consequences and policy implications. The Bangladesh DevelopmentStudies, 22(2/3), 217–243. Retrieved from http://www.jstor.org/stable/40795513

Afsar, R. (2002). Gender dimensions of labour migration in Dhaka city’sformal manufacturing sector. In S. Miller & J. Vivian (Eds.), Women’semployment in the textile manufacturing sectors of Bangladesh andMorocco (pp. 103–150). Switzerland: UNRISD.

Afsar, R. (2003, June 22–24). Internal migration and the developmentnexus: The case of Bangladesh. Presented at The Regional Conferenceon Migration, Development and Pro-Poor Policy Choices in AsiaDhaka, Dhaka.

Ahmed, S., & Fabjer, E. (2009). Engendering adaptation to climate varia-bility in Gujarat, India. Gender & Development, 17(1), 33–50. doi:10.1080/13552070802696896

Amin, S. (1997). The poverty-purdah trap in rural Bangladesh:Implications for women’s role in the family. Development andChange, 28(2), 213–233. doi:10.1111/1467-7660.00041

Ayeb-Karlsson, S., Van der Geest, K., Ahmed, I., Huq, S., & Warner, K.(2016). A people-centred perspective on climate change, environmentalstress, and livelihood resilience in Bangladesh. Sustainability Science, 11(4), 679–694. doi:10.1007/s11625-016-0379-z

Banks, N. (2013). Female employment in Dhaka, Bangladesh:Participation, perceptions and pressures. Environment andUrbanization, 25(1), 95–109. doi:10.1177/0956247813477357

Björnberg, K. E., & Hansson, S. O. (2013). Gendering local climate adap-tation. Local Environment: The International Journal of Justice andSustainability, 18(2), 217–223. doi:10.1080/13549839.2012.729571

Black, R., Bennett, S., Thomas, S., & Beddington, J. (2011). Climate change:Migration as adaptation. Nature, 478, 447–449. doi:10.1038/478477a

Black, R., & Collyer, M. (2014). Populations ‘trapped’ at times of crisis.Forced Migration Review, 45. Retrieved from http://www.fmreview.org/crisis/black-collyer#_edn1

Boyd, M., & Grieco, E. (2003). Women and migration: Incorporating gen-der into international migration theory. The Online Journal of theMigration Policy Institute. Retrieved from http://www.migrationpolicy.org/article/women-and-migration-incorporating-gender-international-migration-theory

Cannon, T. (2002). Gender and climate hazards in Bangladesh. Gender &Development, 10(2), 45–50. doi:10.1080/13552070215906

Chant, S. (Ed.). (1992). Gender and migration in developing countries.London: Belhaven Press.

Chindarkar, N. (2012). Gender and climate change-induced migration:Proposing a framework for analysis. Environmental Research Letters,7, 025601. doi:10.1088/1748-9326/7/2/025601

Curran, S., Shafer, S., Donato, K., & Garip, F. (2006). Mapping gender andmigration in sociological scholarship: Is it segregation or integration?International Migration Review, 40(1), 199–223. doi:10.1111/j.1747-7379.2006.00008.x

Deshingkar, P. (2005, March 14–16). Maximizing the benefits of internalmigration for development. Presented at The Regional Conference onMigration and Development in Asia, Lanzhou.

Evertsen, K. F. (2015). “I lifted an anchor by coming here”. An analysis ofhow gender influences perceived options for adaptive migration in aBangladeshi community (Master’s thesis). Paris School ofInternational Affairs, SciencesPo, Paris. Retrieved from http://www.sciencespo.fr/psia/sites/sciencespo.fr.psia/files/Kathinka20Evertsen20Master27s20Thesis%2027.09.2015.pdf

Foresight. (2011).Migration and global environmental change: Future chal-lenges and opportunities (Final Project Report). The Government Office

for Science, London. Retrieved from https://www.gov.uk/government/publications/migration-and-global-environmental-change-future-challenges-and-opportunities

Fothergill, A. (1996). Gender, risk, and disaster. International Journal ofMass Emergencies and Disasters, 14(1), 33–56. Retrieved from http://www.ijmed.org/articles/96/download/

Gerring, J. (2004). What is a case study and what is it good for? AmericanPolitical Science Review, 98(2), 341–354. Retrieved from http://www.jstor.org/stable/4145316

Gosh, J. (2009). Migration and gender empowerment: Recent trends andemerging issues. United Nations Development Programme HumanDevelopment Reports, Research Paper (Vol. 4). Retrieved from http://hdr.undp.org/sites/default/files/hdrp_2009_04.pdf

Grothmann, T., & Patt, A. (2005). Adaptive capacity and human cognition:The process of individual adaptation to climate change. GlobalEnvironmental Change, 15, 199–213. doi:10.1016/j.gloenvcha.2005.01.002

Huda, S. (2006). Dowry in Bangladesh: Compromising women’s rights.South Asia Research, 26(3), 249–268. doi:10.1177/0262728006071707

Human Rights Watch (HRW). (2015). “Whoever raises their head suffersthe most” workers’ rights in Bangladesh’s garment factories. New York.Retrieved from https://www.hrw.org/report/2015/04/22/whoever-raises-their-head-suffers-most/workers-rights-bangladeshs-garment

Hunter, L. M. (2005). Migration and environmental hazards. Populationand Environment, 26(4), 273–302. doi:10.1007/s11111-005-3343-x

Huq-Hussain, S. (1995). Fighting poverty: The economic adjustment offemale migrants in Dhaka. Environment and Urbanization, 7(2), 51–66. doi:10.1177/095624789500700211

Ikeda, K. (1995). Gender differences in human loss and vulnerability innatural disasters: A case study from Bangladesh. Indian Journal ofGender Studies, 2(2), 171–193. doi:10.1177/097152159500200202

Inmam, M. (2009). Where warming hits hard. Nature Reports ClimateChange. doi:10.1038/climate.2009.3

The Intergovernmental Panel on Climate Change (IPCC). (2014). Climatechange 2014: Impacts, adaptation, and vulnerability. Contribution ofWorking Group II to the Fifth Assessment Report of theIntergovernmental Panel on Climate Change, Cambridge UniversityPress, Cambridge.

International Organization for Migration (IOM). (2009). Gender andlabour migration in Asia. Geneva. Retrieved from http://publications.iom.int/system/files/pdf/gender_and_labour_migration_asia.pdf

Jolly, S., & Reeves, H. (2005). Gender and migration. Overview report,BRIDGE, Institute of Development Studies, Brighton. Retrieved fromhttp://www.bridge.ids.ac.uk/sites/bridge.ids.ac.uk/files/reports/CEP-Mig-OR.pdf

Jones, L., & Boyd, E. (2011). Exploring social barriers to adaptation:Insights from Western Nepal. Global Environmental Change, 21,1262–1274. doi:10.1016/j.gloenvcha.2011.06.002

Jungehülsing, J. (2010). Women who go, women who stay: Reactions to cli-mate change. A case study on migration and gender in Chiapas. Berlin:Henrich Böll Stiftung. Retrieved from https://us.boell.org/sites/default/files/downloads/MIGRACION_Gender_Climate_Mexico_ingles.pdf

Kabeer, N. (1991). Cultural dopes or rational fools? Women and laboursupply in the Bangladesh garment industry. The European Journal ofDevelopment Research, 3(1): 133–160. doi:10.1080/09578819108426544

Kabeer, N., Mahmud, S., & Tasneen, S. (2011). Does paid work provide apathway to women’s empowerment? Empirical findings fromBangladesh. IDS Working Papers Vol 2011(375), Institute ofDevelopment Studies, Brighton. Retrieved from http://www.ids.ac.uk/files/dmfile/Wp375.pdf

Klein, R. J. T., Midgley, G. F., Preston, B. L., Alam, M., Berkhout, F. G. H.,Dow, K., & Shaw, M. R. (2014). Adaptation opportunities, constraints,and limits. In C. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D.Mastrandrea, T. E. Bilir, M. Chatterjee, K. L. Ebi, Y. O. Estrada, R. C.Genova, B. Girma, E. S. Kissel, A. N. Levy, S. MacCracken, P. R.Mastrandrea, & L. L. White (Eds.), Climate change 2014: Impacts, adap-tation, and vulnerability. Part A: Global and sectoral aspects.Contribution of working group II to the fifth assessment report of theintergovernmental panel on climate change (pp. 899–943). Cambridge:Cambridge University Press.

10 K. F. EVERTSEN AND K. VAN DER GEEST

Martin, M., Billah, M., Siddiqui, T., Abrar, C., Black, R., & Kniveton, D.(2014). Climate-related migration in rural Bangladesh: A behaviouralmodel. Population and Environment, 36(1), 85–110. doi:10.1007/s11111-014-0207-2

Martin, S. F. (2003, December 2–4). Women and migration. Presented atConsultative Meeting on Migration and Mobility and How ThisMovement Affects Women, United Nations, Division for theAdvancement of Women, Malmö.

Massey, D. S., Fischer,M. J., & Capoferro, C. (2006). Internationalmigrationand gender in Latin America: A comparative analysis. InternationalMigration, 44(5), 63–91. doi:10.1111/j.1468-2435.2006.00387.x

Mazumdar, M. & Agnihotri, I. (2014). Chapter 7: Traversing myriad trails:Tracking gender and labour migration across India. In T. Troung, D.Gasper, J. Handmaker, & S. Bergh (Eds.). Migration, gender and socialjustice, perspectives on human insecurity, hexagon series on human andenvironmental security and peace (Vol. 9). doi:10.1007/978-3-642-28012-2

McNamara, K. E., Olson, L. L., & Rahman, A. M. (2016). Insecure hope:The challenges faced by urban slum dwellers in Bhola Slum,Bangladesh. Migration and Development, 5(1), 1–15. doi:10.1080/21632324.2015.1082231

Mucke, P. (2014). Urbanization - trends and risk assessment. InWorld riskindex (pp. 53–62). Bonn: United Nations University, Institute forEnvironment and Human Security. Retrieved from https://i.unu.edu/media/ehs.unu.edu/news/4070/11895.pdf

Nelson, V., Meadows, K., Cannon, T., Morton, J., & Martin, A. (2002).Uncertain predictions, invisible impacts, and the need to mainstreamgender in climate change adaptations. Gender and Development, 10(2), 51–59. Retrived from http://www.jstor.org/stable/4030574

Pedraza, S. (1991). Women and migration: The social consequences ofgender. Annual Review of Sociology, 17, 303–325. Retrieved fromhttp://www.jstor.org/stable/2083345

Piguet, E. (2013). From “primitive migration” to “climate refugees”: Thecurious fate of the natural environment in migration studies. Annalsof the Association of American Geographers, 103(1), 148–162. doi:10.1080/00045608.2012.696233

Renaud, F., Bogardi, J., Dun, O., & Warner, K. (2007). Control, adapt, orflee. How to face environmental migration? Interdisciplinary SecurityConnections, Publication Series of UNU-EHS, No 5. Retrieved fromhttp://www.ehs.unu.edu/article:345

Siddiqui, T. (2003, June 22–24). Migration as a livelihood strategy of thepoor: The Bangladesh case. Presented at the Regional Conference onMigration, Development and Pro-Poor Policy Choices in Asia, Dhaka.

Simon, D. (2010). The challenges of global environmental change forUrban Africa. Urban Forum, 21, 235–248. doi:10.1007/s12132-010-9093-6

Sundari, S. (2005). Migration as a livelihood strategy: A gender perspective.Economic and Political Weekly, 40(22/23), 2295–2303.

Tacoli, C. (2009). Crisis or adaptation? Migration and climate change in acontext of high mobility. Environment and Urbanization, 21, 513–525.doi:10.1177/0956247809342182

Tacoli, C., & Mabala, R. (2010). Exploring mobility and migration in thecontext of rural—urban linkages: Why gender and generation matter.Environment and Urbanization, 22(2), 389–395. doi:10.1177/0956247810379935

Van der Geest, K., & Warner, K. (2015). What the IPCC 5th assessmentreport has to say about loss and damage. UNU-EHS Working Paper(Vol. 21). United Nations University, Institute for Environment andHuman Security, Bonn. Retrieved from http://collections.unu.edu/view/UNU:3172

Walsham, M. (2010). Assessing the evidence: Environment, climate changeand migration in Bangladesh. Dhaka: International Organization ofMigration (IOM). Retrieved from https://publications.iom.int/books/assessing-evidence-environment-climate-change-and-migration-bangladesh

Warner, K., Van der Geest, K., & Kreft, S. (2013). Pushed to the limit:Evidence of climate change-related loss and damage when people faceconstrains and limits to adaptation (Report No. 11). United NationsUniversity, Institute for Environment and Human Security, Bonn.Retrieved from http://collections.unu.edu/view/UNU:1849

Yin, R. K. (1994). Case study research. Design and methods (2nd ed., Vol. 5)Applied Social Research Methods Series. London: Sage Publications.

Zlotnik, H. (2003). The global dimensions of female migration. The OnlineJournal of the Migration Policy Institute. Retrieved from http://www.migrationpolicy.org/article/global-dimensions-female-migration

Annex A.

List of respondentsName* Age Occupation/ life situation DateIn Bhola slumMale respondentsAbdul 35 Day labourer 01.09.2014Rafiq 33 Sanitary craftsman 01.09.2014Malik 20 Day labourer 02.09.2014Amir 25 Day labourer 03.09.2014Tasfiq 23 Van driver 13.09.2014Faisul 35 Day labourer 13.09.2014Reaj 35 Day labourer 15.09.2014Focus group discussionHafiz Over 60 Retired day labourer 17.09.2014Bashir 40 Day labourer 17.09.2014Ahmad 40 Van puller 17.09.2014Rahman 40 Rickshaw puller 17.09.2014Habib 32 Construction worker 17.09.2014Salim 40 Shop owner 17.09.2014Female respondentsAnika 17 Garment worker 29.08.2014Israt 18 Garment worker 29.08.2014Seema 19 Garment worker 29.08.2014Rayhana 25 Garment worker, just lost her job 04.09.2014Nahar 35 Housemaid/tea-vendor 04.09.2014Ayela 25 Housemaid 05.09.2014Adila 31 Housemaid 05.09.2014Farzana uknown Garment worker 05.09.2014Samira 20 Looking for work 11.09.2014Nadia 18 Garment worker 11.09.2014Nyala 35 Previous housemaid and garment worker 11.09.2014Nasrin 16 Garment worker 12.09.2014Naureen 17 Garment worker 12.09.2014In Bhola districtMale respondentsNasir 28 Fisherman 25.09.2014Sharif Over 60 Retired fisherman 27.09.2014Focus group discussionRasul Fisherman 29.09.2014Zarid Fisherman 29.09.2014Nasif Fisherman 29.09.2014Alam Fisherman 29.09.2014Fahim Fisherman 29.09.2014Kamal Fisherman 29.09.2014Female respondentsFarhana 35 Housewife 24.09.2014Halima 50 Mother of garment worker 24.09.2014Faiza 50 Mother of garment worker 25.09.2015Seleha 24 Returned garment worker 25.09.2015Focus group discussionJameela 19–35 Housewife, returned garment worker 26.09.2014Nasreen 19–35 Housewife, returned garment worker 26.09.2014Sabah 19–35 Housewife, returned garment worker 26.09.2014Fatima 19–35 Housewife, returned garment worker 26.09.2014Shirin 19–35 Housewife, returned garment worker 26.09.2014Naju 19–35 Housewife, returned garment worker 26.09.2014Aysha 19–35 Housewife, returned garment worker 26.09.2014Tahera 19–35 Housewife, returned garment worker 26.09.2014Focus group discussionRehana 35 Housewife 27–28.09.2014Sabiha 35 Housewife 27–28.09.2014Nasrat 25 Housewife 27–28.09.2014Rumana 50 Housewife 27–28.09.2014Nyala 30 Housewife 27–28.09.2014Soraya 35 Housewife 27–28.09.2014

*All names have been changed to maintain respondents’ privacy.

CLIMATE AND DEVELOPMENT 11